Adopting a new sine-induced statistical model and deep learning methods for the empirical exploration of the music and reliability data

Abstract

The presence of probability-driven models is highly influential in setting the stage for vital decision-making in domains including reliability, engineering, music engineering, and other closely interconnected scenarios. With a deep understanding of the consequential roles played by probability-arisen models, we have developed and implemented a new probabilistic model. This model is constructed by utilizing the sine-based function and the exponentiated Weibull distribution, and it is known as the exponent power sine exponentiated Weibull (EPSE-Weibull) distribution. Point estimators are derived for the EPSE-Weibull distribution. These estimators are then evaluated through a simulation study. The significance of the EPSE-Weibull distribution is demonstrated through the analysis of reliability and music engineering data sets. In addition to the above, we also utilize two deep learning algorithms, namely Artificial Neural Networks (ANN) and Support Vector Regression (SVR), to forecast the same data sets. The findings indicate that the ANN model consistently exhibits higher levels of accuracy, as evidenced by its lower Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) values compared to the SVR model for both data sets. These findings indicate that ANN is better at capturing the fundamental patterns in the underlying data sets. In addition, visual representations, such as bar charts and line charts, further emphasize the superior performance of the ANN across both data sets.

Keywords

• Weibull distribution
• Exponentiated weibull distribution
• Sine function
• Reliability
• Music engineering
• Statistical modeling